802.11ax WiFi: Promising, but Challenging

By now, you’ve likely heard of 802.11ax and the hype about it being the next big thing in WiFi. There’s certainly a lot to be excited about, and the big performance numbers along with new features will have many marketing folks absolutely giddy. But as with any new technology, there is wheat to separate from the chaff, and a number of concerns for those of us who design and manage wireless networks.

For those needing a little catch-up, 802.11ax is the pending successor to the 802.11ac WiFi standard. It’s expected to be ratified in late 2019 as a formality, but we’re already seeing pre-standard products in both the consumer and enterprise WLAN markets. The short version of what it brings: better, faster WiFi in both 2.4 GHz and 5 GHz and a raft of new features and lexicon to the WLAN world. There's a wealth of information about .11ax, including Aerohive’s FAQ, Cisco’s 802.11ax white paper, and a nice primer from National Instruments. However, sexy performance promises are one thing; getting there is a whole other conversation.

Backwards compatibility vs. innovation

As 802.11ax gains steam, we’ll hear vendors tout stratospheric WiFi data rates measured in gigabits and how access points need crazy fat uplinks -- up to 10 Gbps, depending on the vendor's motivation to sell switches. But .11ax is also about backwards compatibility all the way back to 802.11b, where data rates of 1 and 2 Mbps are found! In environments with a lot of 802.11a/b/g and even early 802.11n clients, .11ax may be somewhat pointless unless you expect client updates. The IEEE refuses to let go of the past when it comes to WLAN innovation, and old clients slow fast networks down. You can’t have it both ways over WiFi.

Underlying code

When it comes to 802.11ax -- or any wireless technology -- performance and reliability will only be as good as the underlying code. With a slew of new features and radio magic, .11ax will challenge WLAN coders big time. At the same time, certain market-leading WLAN vendors already struggle mightily with wireless code bugs, even as they try to hook trendy technologies like SDN, AI, and automation features into already frustrating WLAN firmware. This is one of the top anxieties for those of us who handle large-scale WLAN environments when we look ahead to .11ax.

When do you make the leap?

This question isn’t .11ax-specific, but it can have fresh implications in the context of the new WLAN standard. The truth is that many environments now running pricey .11ac gear would have been fine if they had never migrated from .11n. That’s not to say that .11ax won’t be beneficial to many wireless networks, but the benefits will depend on a healthy number of .11ax client devices. Several industries including healthcare and retail are notorious for keeping old or limited wireless clients, and that’s not likely to change. The siren song of .11ax will be strong, but when the bulk of your devices are 802.11g retail scanners, the technology may not be the best use of precious IT dollars.

WiFi.jpg

Even where .11ax does make sense, the difficulty in migration will be a matter of scale. While SMB environments may have just a few access points to change out, large enterprises that count their APs in the hundreds or thousands may also be looking at new switch costs (if PoE isn’t up to snuff or multi-gigabit is desired), new wireless controllers, and even new wiring costs depending on WLAN design.

WLAN design considerations

Every new WLAN technology deployment that replaces the last should have a fresh WLAN survey and at least some thought given to whether the old design will work for new APs. At the same time, re-using old AP wiring runs can be a hard temptation to resist. Further complicating matters, design toolmakers like iBwave and Ekahau need to be fully invested in .11ax -- and the surveying engineer needs to have an .11ax client device -- to make an upgrade valid. Until we get there, it will be hard to bring real accuracy to .11ax designs and verification surveys.

Too much hype

To this day, we see 802.11ac oversold on what it can really do. Maybe marketers don’t quite get technical reality or maybe reporters covering wireless just don’t dig deep enough at times. Regardless, as much as .11ac is overhyped at times, .11ax will be brutal in this regard. It comes with the territory, but it also leads to sales folk making promises that network admins just can’t deliver, and frustration all around. For WLAN professionals, staying above it all means having a solid command of what .11ax is and is not, both in general and in our own specific environments, and being willing to educate everyone from end users to the CIO about the reality.

At the same time, I'm excited about the promise of .11ax. I’ve been in the WiFi game for many years and can say that every new WLAN technology creates excitement, some degree of frustration, and most importantly, better wireless networks when designed and administered by skilled hands. This will hold true with .11ax.  But we need to move forward with eyes open and the right amount of skepticism.